MkIV Column Results during MOHAVE 2009

Measurements Of Humidity in the Atmosphere and Validation Experiment

Most comprehensive H2O intercomparison to date:

•3 water vapor Raman Lidars: nighttime measurements: ground-20 km

•15 cyrogenic frostpoint hygrometers launched: ground-30 km

•3 frost-point hygrometer radiosondes launched: ground-30 km.

•50 Vaisala RS92 radiosondes launched: ground-15 km.

•2 microwave radiometers: night and day, vertical range: 20-80 km.

•2 GPS receivers: night and day, total column H2O

•2 FTS (ground-based, solar): MkIV, FTUVS

•Satellite observations by MIPAS, TES, AIRS and ACE.

Covered a two-week period October 12-28, 2009.

Organized by Thierry Leblanc at JPL’s Table Mountain Facility (2.26 km)

http://tmf-lidar.jpl.nasa.gov/campaigns/mohave2009.htm

Special issue of AMT – deadline May 2011

MOHAVE-2009 Campaign

http://tmf-lidar.jpl.nasa.gov/campaigns/mohave2009.htmhttp://tmf-lidar.jpl.nasa.gov/campaigns/mohave2009.htmhttp://tmf-lidar.jpl.nasa.gov/campaigns/mohave2009.htm

JPL MkIV Involvement

MkIV Interferometer:

• Primarily a balloon instrument

• Records entire 650-5650 cm-1 spectral region simultaneously (parallel

HgCdTe and InSb detectors) at 0.005 cm-1 resolution (117 cm OPD)

During MOHAVE 2009 Campaign

• Made measurements from Oct 16-28 inclusive (Day 289 to 301).

• Data available on all days except two (one cloudy; one operator error)

Compare MkIV H2O with:

• Coincident Sonde profiles on 7 days

• Co-incident GPS column H2O

• Co-incident FTUVS column H2O

This presentation will focus on column H2O (Matthias Schneider has already

published two papers on H2O profiles retrieval from MkIV/MOHAVE-2009)

Also look at other gases and trajectory analyses for context.

Spectral Analysis

Spectral fitting retrievals performed by GFIT (profile scaling)

Use NCEP analyses for P/T/H2O profiles

Used MkIV balloon climatology for other gases

Used low-pressure HBr and OCS cells for ILS calibration

Center Width Gases Fitted SMAX E” Bias SE 2/N

841.90 1.20 h2o 2.445E-24 601 0.9156 0.0006 0.471

1111.5 3.00 h2o o3 3.37E-23 975 1.0654 0.0006 5.09

1117.6 0.70 h2o o3 2.88E-24 2206 1.0560 0.0008 3.05

1121.1 2.00 h2o o3 5.61E-23 1263 1.0605 0.0009 3.48

1959.65 0.70 h2o 6.213E-24 451 1.0634 0.0012 0.467

2819.45 2.60 h2o ch4 n2o hcl 6.616E-24 783 1.0962 0.0009 1.022

2871.10 2.20 h2o ch4 hdo 2.368E-24 504 1.0794 0.0009 0.394

3001.20 0.65 h2o ch4 1.182E-24 540 0.9838 0.0017 0.473

3019.85 0.70 h2o o3 hdo 1.872E-24 423 0.9882 0.0007 0.588

3155.21 1.02 h2o ch4 4.962E-24 449 1.0908 0.0006 0.494

3205.45 1.20 h2o 2.501E-24 173 0.9771 0.0004 0.827

4056.54 0.46 h2o 5.850E-25 394 0.9054 0.0007 1.184

4537.29 1.50 h2o ch4 2.755E-24 562 1.0256 0.0008 0.573

4543.01 0.72 h2o ch4 6.923E-25 384 0.9749 0.0011 0.659

4552.92 2.56 h2o ch4 1.364E-24 681 1.0214 0.0008 0.640

4556.21 2.50 h2o ch4 3.800E-24 369 1.0001 0.0009 0.278

4565.20 2.50 h2o co2 ch4 3.627E-24 342 1.0111 0.0007 0.244

4571.05 1.01 h2o co2 ch4 1.234E-24 240 1.0498 0.0010 0.563

4576.85 1.90 h2o ch4 1.178E-23 356 1.0277 0.0011 0.833

4598.69 10.78 h2o ch4 co2 n2o 3.865E-24 329 1.0162 0.0007 0.180

4611.05 2.20 h2o ch4 co2 n2o 6.253E-24 80 1.0028 0.0007 0.417

4622.00 2.30 h2o co2 n2o 8.754E-24 217 1.0244 0.0007 0.570

4631.55 1.40 h2o 2.856E-24 176 0.9491 0.0007 0.685

4699.55 4.00 h2o co2 n2o 6.858E-24 25 1.0268 0.0007 0.533

4734.60 7.30 h2o co2 3.336E-24 325 0.9879 0.0008 0.402

4761.15 10.70 h2o co2 6.818E-24 305 1.0122 0.0008 0.530

Fitted H2O Windows (2)

The wide simultaneous coverage of the MkIV instrument allows comparison of

H2O windows in different spectral regions.

The large range (factor 20) of available line strengths provide high dynamic range

The best H216O windows are mostly at higher frequencies (> 4500 cm-1). At lower

frequencies, the T-insensitive H216O lines tend to be saturated

Biases vary from -10% to +10% from window to window, but are generally 3%

Small values of 2/N imply non-varying systematic residuals which drive up

computed uncertainty but do not degrade spectrum-to-spectrum precision

Large values of 2/N imply poor spectrum-to-spectrum precision, despite good

fitting residuals (varying systematic errors that don’t degrade fits)

2/N values tend to be largest for T-sensitive windows (because I do not perform

a simultaneous temperature retrieval).

MkIV – FTUVS Comparison•Fourier Transform Ultra-Violet Spectrometer (FTUVS) instrument is

permanently installed at TMF. 10 cm OPD (0.06 cm-1 resolution)

•Its main purpose is measurement of OH at 308 nm and NO2 in the visible.

•For MOHAVE-2009, instrument was adapted to cover the SWIR region

where eleven H2O windows were used covering 6076-6470 cm-1.

MkIV – FTUVS Comparison

Data color-coded by day of year:

Dark Blue = Day 289

Red = Day 301

Correlation is good apart from

day 289 on which the FTUVS

to MkIV ratio was 5% low as

compared with other days.

FTS – GPS-TABV comparison

MkIV – GPS-TABV Correlation

Color-coded by date:

Purple= Day 289 (Oct 16)

Red= Day 301 (Oct 28)

MkIV PWV correlation with GPS-TABV

is poorer than with FTUVS.

This should not be surprising: MkIV and

FTUVS are very similar techniques:

•direct solar absorption

•same data analysis algorithm

•same spectroscopic database

GPS is a completely different technique

GPS PWV seems to level off at low H2O

but for large H2O the gradient is ~1

IPW Comparison Summary

Depending on the retrieval method and H2O lines used,

estimates of IPW from MkIV spectra vary by up to 5%.

Apart from day 289, MkIV-FTUVS are in surprisingly good

agreement 1.020.03 (common GFIT analysis) given that a

completely different set of H2O lines was used.

GPS-TABV data are in good agreement with FTS at

IPW>2mm, but deviate at lower values. Deviation is in

opposite direction to that reported by Schneider et al. [2010]

O3 Time Series – PV correlation

HF Time Series – PV correlation

N2O- Time Series – PV Correlation

Los Angeles

San Francisco

H2O

HDO/H2O

H2O and HDO variation

Los Angeles

San Francisco

CO

C2H6

C2H2

MkIV measurements of CO, C2H6, and

C2H2 show no obvious correlation with

48h HYSPLIT back-trajectories.

Only one day (291) has track over LA

and it has lower than normal amounts

Pollution from LA is clearly not a factor

Day of Year

Measurements of Pollutants

H2CO

HCOOH

NH3

Los Angeles

San Francisco

Day of Year

Shorter-Lived Pollutants

Summary of Other Gases

Column N2O and CH4 correlates with tropopause altitude.

Column O3 anti-correlates very well with tropopause altitude

Column HF anti-correlates less well with tropopause altitude.

Column H2O correlates with tropopause altitude, but the correlation is

probably indirect:

low tropopause polar airmass cold temperatures low H2O

Little correlation between: H2O , pollutants, and 48h trajectories

Given the close proximity of TMF to LA, it is surprising how little affected

by LA pollution the measurements were. The one day (291) that the

trajectory crossed the LA basin was among the cleanest.

Stratospheric Gases